The battle for AI talent: Where have all the fresh graduates earning million-yuan annual salaries gone?

“I started sending out resumes in early August 2025. After multiple rounds of interviews, I received offers for algorithm positions from several leading tech giants such as ByteDance, Alibaba Cloud, and Tencent.” Yu Bai (pseudonym), a master's graduate from the Department of Electronic Engineering at Tsinghua University, was extremely interested in the application and innovation of large AI models. In the end, he accepted an offer from Meituan's “Beidou Plan” during the 2026 campus recruitment. He will officially start work in early July.

With the rapid development of AI, the demand for talent in popular positions such as AI engineers, data engineers, and chip engineers has been continuously expanding since this year. At the same time, when numerous outstanding job seekers compete for an SSP offer (Super Special Offer), it's like thousands of troops crossing a single-plank bridge. Yu Bai is a microcosm of this situation.

What do tech giants value in recruitment?

Yu Bai studied communication engineering at Jilin University as an undergraduate. During his postgraduate studies at Tsinghua University, he engaged in AI-related research under the guidance of his tutor. He told China News Service that before officially participating in campus recruitment, he had already accumulated internship experiences in large model-related positions at several companies, including a cutting-edge technology company, AMD, and Tencent.

Looking back on the process of obtaining offers from multiple tech giants, Yu Bai summarized that the core experience lies in thinking while reading cutting-edge papers and taking practical actions in specific projects.

“During the interview, what is more concerned about is whether the candidate can abstract the problems in the business scenario into topics that can be solved by AI technology.” Yu Bai recalled that during a previous interview with Gaode, the interviewer proposed a problem of automatic tourist route recommendation. He suggested combining AI with path planning and quantifying subjective content into objective indicators through retrieval technology. Yu Bai thought the interviewer gave a positive feedback on this idea.

Yu Bai noticed that with the upgrade of AI-assisted tools, as long as one can use AI tools, they can quickly finish writing code. Nowadays, for large model algorithm positions, candidates are required to have not only knowledge reserves and programming abilities. “Interviewers value your understanding and cognition of the entire technology, especially systematic views in vertical fields.”

Cheng Shui (pseudonym), who also received an offer from Meituan's 2026 campus recruitment, mentioned that tech giants especially value systematic and mechanism-based innovation and collaboration abilities. During her internship at Meituan, Cheng Shui was responsible for managing the information distribution system in the work area. She actively summarized a set of management processes and made a mechanism table, which surprised her supervisor. Cheng Shui summarized that three factors mainly contributed to her successful full-time conversion: outstanding personal ability, cross-department communication and collaboration ability, and the initiative to create value for the enterprise.

Positions of AI, data, and chip engineers lead in growth rate

According to a recent report released by Zhaopin, the Internet industry ranked first in the list of the top 10 industries with the highest demand for fresh graduates from January to May 2026. From a professional perspective, AI engineers ranked first in the growth rate of demand for fresh graduates with a year-on-year growth rate of 31.1% in the number of positions. Data engineers and chip engineers followed with growth rates of 28.3% and 11.1% respectively.

According to Yu Bai, the annual salary packages offered by several Internet companies for large model algorithm-related positions that gave him offers exceeded one million yuan.

Image source: Zhaopin

Zhaopin data shows that from January to May 2026, in the Internet industry, the positions with relatively fast growth rates in the number of recruitment positions in enterprises with more than 1,000 employees included chip engineers, mechanical design/manufacturing engineers, electronic/electrical/automation engineers, AI engineers, data engineers, communication and hardware R & D, and operation and maintenance support. Among them, the growth rate of the number of recruitment positions for chip engineers in enterprises reached 60%.

Data source: Zhaopin

Li Chang'an, a researcher at the National Institute of Opening-up at the University of International Business and Economics, told China News Service that currently, the demand for technical positions in the field of artificial intelligence continues to skyrocket. Professional talents in areas such as algorithms and chips have become the core resources that major tech giants are vying for. With the deepening of AI application scenarios, enterprises not only need technology developers but also composite talents who can proficiently use AI tools to complete instruction writing and content creation. “New liberal arts students” are presented with opportunities. For example, positions such as short drama creation, AI model instruction design, and scenario script writing require the excellent expression, organization, and situation description abilities that liberal arts students possess.

Li Chang'an believes that currently, leading Internet companies value the comprehensive suitability of job seekers more in recruitment. Applicants should not only have a certain ability to understand technology but also possess flexible application abilities and innovative thinking. For fresh graduates, combining their original professional advantages with AI tools in depth will give them more differentiated competitiveness.

“Do the most core and cutting-edge things in the AI era”

Recently, Baidu launched the 2027 AIDU Program (Baidu's top AI talent recruitment program), which is open to doctoral students and outstanding master's students of the class of 2027.

At the Baidu Scholars sub-forum of the Create2026 Conference held in May, He Junjie, a senior vice president of Baidu Group, revealed that a 2025 campus recruit at Baidu has now grown into a product expert in the field of digital human agents. He Junjie said that this is a direct manifestation of Baidu's concept of “letting young people take on major responsibilities,” allowing young talents to directly enter the business battlefield and do the most core and cutting-edge things in the AI era.

As early as March this year, Alibaba also launched its 2027 intern recruitment. Alibaba said that more than 80% of the positions in this campus recruitment are related to AI, and seven new types of AI campus recruitment positions have been added, including AI application R & D engineers, AI application algorithm engineers, and AI Agent optimization engineers. New employees will be deeply involved in the research of multimodal AI topics after joining the company.

The 2027 AliStar top talent recruitment and training program was also launched simultaneously, covering cutting-edge topics in five major fields: basic models, AI Infra, large model applications, industrial AI, and computing architecture.

According to a relevant person in charge of Meituan's recruitment, the “Beidou Plan,” which recruits top campus technology talents globally, is open for recruitment throughout the year. This plan focuses on cutting-edge technology directions such as large model bases, large model applications, autonomous driving, drones, and intelligent decision-making.

The views in this article are for reference only and do not constitute investment advice. Investment is risky, and you should be cautious when entering the market.

This article is from the WeChat official account “China News Service”. Author: Xie Jingwen. Republished by 36Kr with authorization.